Knee-Braced Frames

[enigma2]

We have a low rise building with 16’-0” floor to floor heights. We had intended to utilize braced frames for the lateral force resisting system. However, in working with the owner, we could not agree on a location to obstruct tenant space. Going to moment frames is cost prohibitive. Since there is a large space between the ceilings and floors, we have opted to provide steel knee-braced frames so all the bracing is hidden in the ceiling space. Wind loads control the design over seismic loads.

We’ve had a number of “what if?” debates in the office about this framing scheme:

What if seismic controlled?

In this case, what type of frame would this be? The ASCE 7-05 definition of an EBF states “a diagonally braced frame in which at least one end of each brace frames into a beam a short distance from a beam-column or from another diagonal brace.” If one end of the brace is also framing into the column, is at an EBF? If not an EBF, what is it?

Is a knee-braced frame an acceptable seismic-force resisting system?

If it is not an EBF, is the only category it can fall into the “Steel Systems Not Specifically Detailed for Seismic Resistance…” per ASCE 7-05? If we’re in SDC A, B or C, we almost exclusively use R=3 and this SFRS. Are there any good resources out there that really spell out when this can or cannot be used?

 

[msquared48]

Why not avoid the problem and just go with an Ordinary Moment Frame with W shapes?

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[JoshPlumSE]

It wouldn't be an eccentrically braced frame. Those require extesive detailing of the link section of the beam or column.

I think it is pretty clearly one of the "steel systems not specifically detailed for seismic resistance" with an R <=3.0.

Now, the question just becomes "what codes govern when you are allowed to use this type of system." I can't say that I've looked at that much... I always use systems with R > 3.0. But, that's probably because I'm in California.

 

[WillisV]

I agree with Charlie:

http://www.seaintarchive.org/group/seaoc/mailarchive/2002d/msg01654.html

 

[JAE]

I would also lean to what WillisV linked to.

I would also at least mildly challenge the term "wind loads control the design".

In many cases, the total lateral load force from wind is greater than that from seismic. But in a true sense, ALL the load combinations of the governing code must be satisfied and many times you will find that some bits and pieces of the structure are governed by the seismic portions of the code vs. the wind portions, even when the overall wind force is greater.

You cannot just determine the overall wind force, the overall seismic force, compare them, and then disregard the smaller type of loading from then going forward. The seismic requirements still are valid even if the wind load globally is larger.

So if there are seismic code limitations on the use of a K framing concept, whether wind or seismic "governs" doesn't matter. Those limitations still apply.

 

[enigma2]

I found in AISC 360 that knee braces should be treated as moment frames since they essentially function in that regard. The Seismic Provisions consider a knee-brace system as an OMF. This would apply if we were using anything other than R=3. I agree it must fall under the "steel systems not specifically detailed for seismic resistance" scheme.

On a tangent: In the case of this particular project, in examining all of the load combinations, there was not one in which a combination with seismic loads governed on any one of the structure's members. While JAE's statement is correct and good practice, in this case, wind truly does govern.

 

[JoshPlumSE]

I'm not challenging enigma's claim. But, I always like to comment when someone says that "wind governs the design".

You can get into cases where the overall wind shears are significantly higher than the seismic shears. But remember, those seismic shears are reduced based on the assumed ductility and energy dissipation of the structure. For that to happen, you will often need to design your connections to a much higher force.... Meaning that Seismic can still govern the design of the connections.

I find that this is a sublety that is often lost on young engineers in California. Or, even with engineers from low seismic regions.

 

[trainguy]

I think about 40 yrs ago, this was an often stated argument. Engineers would simply compare the 2 total loads on the building, choose 1 and go forward.

I think we now know more about ductility, assumed plasticity, elastic vs plastic response etc.

Completely agree with JAE & JoshPlum.

tg

 

[ajose]

I'm agree with the answer given by JoshPlum, JAE & Trainguy about the bracing, and i would like to add:

Your detaling cannot be qualified like EBF, it is just a knee bracing. To qualified like EBF, the bracing goes from node beam-colum to beam or beam to beam. And this is for guarantee the plastic behavior otuside of the column. And the code give all the provisions needed to keep the plastic hinge outside the column (on the beam).

The forces from seismic analisys have an handicap due structure conception (do not fit with seismic code). You can try to go bye using a proper seismic analisys. The seismic "R" factor is function of the ductility of the building or in other words the energy that can be dissipated. So if you made an analisys with the bracing like you stated, have to use R=1 to make sure elastic behavior on all the structure and no need to inelastic behavior. It's expensive but place on real context seismic and wind forces. You have to move de designing to ASD (AISC360-05).

If the new seismic analisys there has no problem, then i see no problem. For shure you are not going to need ductility becuase you are on just elastic behavior. You have to review is there is no additional requirements on codes, i can not remember one.

This way of thinking make more expensive the structure.

Other item to check is the total hight of the structure, ¿the seismic code allow not to use special detailing?

ajose